benchmark and competitive analysis of port performances...

Ouail EL IMRANI and Aziz BABOUNIA, International Journal of Research in Management, Economics and

Commerce, ISSN 2250-057X, Impact Factor: 6.384, Volume 08 Issue 06, June 2018, Page 36-49

http://indusedu.org Page 36

This work is licensed under a Creative Commons Attribution 4.0 International License

Benchmark and Competitive Analysis of

Port Performances Model: Algeciras Bay,

Rotterdam, New York-New Jersey and

Tangier Med

Ouail EL IMRANI1 and Aziz BABOUNIA

2

1(Pursuing Ph.D. at the University Abdelmalek Essaadi of Tetouan - Morocco, Research Laboratory:

Management, Logistics, Governance and Applied Economics)

2(Research Professor, National School of Commerce and Management - Kenitra, University Ibn Tofail –

Morocco)

Abstract: In this research paper four sea ports namely, Tanger Med, Algeciras Bay, Rotterdam and New York-

New Jersey has been taken into study to understand and evaluate their efficiency of operations and benchmark

them. Port efficiency is the measure of amount of input and output and their ratio. Port efficiency is not solely

dependent on port performance. The port performance strategies of the case ports were studied and efficiency

variables were found through various literatures. To analyze input and output variables of the ports, efficiency

software named Data Envelopment Analysis Program was used to find the most efficient ports. Then the

variables for the most efficient ports were benchmarked and ranked. A hypothetical port efficiency model has

also been suggested for better efficiency of the ports, especially for the maritime lines of the Tanger Med port.

Keywords: ports, efficiency, benchmark, performance, DEA

I. INTRODUCTION Port Efficiency analysis

Port efficiency analysis is the methodology or the technique used to measure the ratio of input and

output of a port(Yang et al., 2011). Port plays a major role in a country’s economy and development by

providing international trade link, thus their efficiency is vital (Liu, 2010). Port efficiency is an important

contributor to an international competitiveness and thus is checked on key performance indicators

(KPIs)(UNCAD, 2016). A port need not be efficient alone, but also it needs to be effective in high and quality

throughput given to customers; shipper, ship-owners, and carriers.The relationship between demand of port’s

throughput services and the port process are known as the port’s throughput demand function which also

influence the port efficiency.There are various factors that affect the efficiency of the ports(Nyema, 2014) like;

capital investments, operational services (towing, pilotage, mooring and others), customs clearing time, financial

and other vessel operations (avg. turnaround time, avg. vessel calls etc.). Analyzing these factors helps to

evaluate the input-output ratio of a particular port.

Container throughput and facility productivity is the main measure and indicator of port performance

for every Seaports (Babounia & EL Imrani., 2016). The selection andchoice of the ports by freight carriers are

usually based on the feasibility of the sea route, port rates, hinterland connectivity, port infrastructure and port

capacity(Ruto & Datche, 2015). Efficient port governance is thereby very important in driving the decisions of

the freight carriers which in turn will help the ports to perform well. Physical quantities of items, scale or scope

of activities, levels of effort expended and the efficiency in converting resources into some kind of product

evaluates the efficiency of a Port. In most of the private corporations the outcome measures are evaluated on

finances and asset utilization(Mpogolo, 2013). Efficiency focused ports measure their performance on the basis

of financial statement, marketing activities of comparison between years and competitors and other similar

strategies with the intentions of expanding gross margin (Bozuwa et al., 2012).

Justification, Significance of the study and Need of the study

Over the last two decades, the global container industry has experienced an annual increase of about 7-

9% with a growth rate of 2% per year (Vacca, Bierlaire and Salani, 2007).

This study attempts to improve the performance of the various ports, especially that of Tangier Med,

and it can be expected that the study will help the various stakeholders to formulate and implement measures

aimed at improving the performance port. On the academic level, research contributes to the implementation of

a benchmarking model adapted to specific situations. The study also contributes significantly to the literature on

port performance measures.





Overview of Case Ports

Port of Tanger Med is located in the African country of Morocco is the newest port which opened in

FY2007 and has a very strategic location at the slopes of Strait of Gibraltar and the Moroccan coast is nearest to

the Iberian Peninsula (Tanger Med Port Authority, 2014). Average ports of call for Africa and Senator Lines are

51 counts(Tanger Med Port Authority, 2016a). The port is committed to quality and environmental approach in

maintaining the performance of the port.The port has been reported to handle containers of 0.6million TEU and

an operational income of $527,906 for the FY2015. For the FY2016 the total tonnage has been reported to

4.4million TEU(Tanger Med Port Authority, 2016a). Recently, to improve the efficiency of the port, the port

authorities have agreed to installation of 68 new industrial projects in Tanger Med complex, along with plans for

private investments of $939 million and create 6,547 new jobs(Tanger Med Port Authority, 2016b).These

projects will include 13 new industrial units in the automotive sector, 11 units in textile, 2 units in aeronautics, 9

logistics projects and 24 SME (micro, small and medium-sized enterprises) projects.

Port of Algeciras Bay is the oldest and EU’s busiest port and is located at Algeciras Bay and Tarifa, at

the southernmost Spain. Port of Algeciras Bay is located between the Atlantic Ocean and the Mediterranean

Seais connected to West Africa, Asia, North Europe, and the Americas. Thisport is known for its natural shelter

and deep-draft conditions which attracts seafarers (Port of Algeciras Bay, 2015). Average ports of call for APL,

Msersk Line and Zim are 160 counts. It was recorded that the total cargo handling for the year 2015 was

4.52million TEUs and a net profit of $941,905(Port of Algeciras Bay, 2016). Port of Algeciras Bay is mainly a

transport hub and industrial center along with fishing industry.They have mainly strengthened the port’s

logistics and operations by cost reductions, increased port capacity, maritime services and most importantly

hinterland connectivity. Moreover, they plan to develop a rolling motorway project to develop a new intermodal

service across the Strait of Gibraltar and linking the EU with the North of Africa(Goodwin, 2015). The port

authorities plan to achieve and handle more than 5 million TEUs by FY2020. In 2015, the Port of Algeciras

received 28,446 ship calls with both the terminals serving more than 3,000 containerships very efficiently(Port

of Algeciras Bay, 2016).

Port of Rotterdam, is Europe’s largest sea port, which is located in Netherlands has daily routes to both

the TangerMed and Algeciras Bay. The Port of Rotterdam has a busy petrochemical industry and several oil

refineries with 209million tonnes of liquid bulk as reported in the FY2015(Port of Rotterdam, 2016a). Crude oil

arrives by sea to be processed and delivered to areas in The Netherlands, Germany, and Belgium. The port has

also been reported to have handled 461.2 million tonnes in total for the FY2015(Port of Rotterdam, 2016c). The

port also handles energy, bio-based, LNG, break bulk, refining and chemicals related cargo. The average port of

call for 33 carriers is 72 counts. It has been analyzed that, every year almost 30,000 sea route carrier and

110,000 inland carriers(Port of Rotterdam, 2016a).The port also aims torise and handle over 15 million TEUs by

FY2020 and also to remain the leading port in Europe(Port of Rotterdam, 2016c).

Port of New York and New Jersey is a major port in the USA which is well connected through major

sea routes to the other three case ports.The combined port together covers an area of 40,000 meters and consists

of small hinterland ports; “Port Newark, the Howland Hook Marine Terminal, the Red Hook Container

Terminal, the Elizabeth Port Authority Marine Terminal”. The port also has 3 cruise terminals for passengers. It

was reported that the port had handled 112million tonnes of dry bulk and annual container handled about

3.7million TEU for the FY2015(PONYNJ, 2016). The average port of call for 18 carriers is 62 counts. The

Council on Port performance maintains and looks after the port performance and manages them. The council

mainly focuses on four main objectives of improvement; equipment, rail, operations and customer care which is

reversibly linked by outreaching the government and the community (PONYNJ, 2014).

Table 1: Summary of Case Ports

Features Port of Tanger Med Port of Algeciras Bay Port of Rotterdam Port of New York-New

Jersey

Opening Date 2007 1906 1962 1921

Location

Morocco (Slopes of

Strait of Gibraltar and

the Moroccan coast is

nearest to the Iberian

Peninsula)

Algeciras Bay and Tarifa,

at the southernmost Spain

Netherlands, near

German Ruhr district,

Paris and London

New York and New

Jersey

Trade Routes Europe, the Americas,

Asia and rest of Africa

West Africa, Asia, North

Europe, and the Americas

The Netherlands,

Germany, and

Belgium

Major sea routes and

connectivity with three

case ports as well

Ports of Call 51 160 72 62

Complex

Includes

Tangier Med 1, 2,

Passenger’s port and

TMPC center

Container terminals, Bulk

and break bulk terminals,

oil/liquid terminals,

Ro/Ro terminals and

Container, Oil,

Break bulk, Large

Parking Area for

transport vehicles,

Passenger terminal,

Cargo Airport, Cargo

Railway, Container and

oil terminals


Commerce, ISSN 2250-057X, Impact Factor: 6.384, Volume 08 Issue 06, June 2018, 6-49



passenger terminals Passenger terminal,

Cargo Railway

Handling

Capacity

4.4 million TEU 4.52million TEUs 461.2 million tonnes

112 million tonnes of

dry bulk and annual

container handled about

3.7 million TEU

Objectives of the Study

The objectives of this research paper are;

a. To investigate the different key performance indicators of the 4 case ports and conduct port efficiency

analysis , by measuring the efficiency of the port and maritime operations of each port, through the

analysis of the elements necessary for the performance such as: the ontrasting Port Attributes, Average

Turn-around Time of Ships, Average vessel call, Tonnage handled.

b. To benchmark ports on different parameters and determine which port is efficient in which area,

through the analysis of the following elements: Input Orientation Efficiency versus Average Ship

Turnaround Time, Input Slip on Average Ship Turnaround Time, Input Orientation Efficiency versus

Average Ship Wait Time and Input Slips, Input Orientation Efficiency versus Average Vessel Calls and

Input Slacks, Input oriented efficiency against Container tonnage and input slacks, Input oriented

efficiency against Container tonnage and input slacks and Input oriented efficiency against vessel and

input slacks of each port.

II. METHODOLOGY Key Performance Indicators

The port’s performance is evaluated by checking if actual throughput overcomes optimum throughput

(Vitsounis, 2012). Port performance indicators are divided into four categories namely; “Ship Operations,

Cargo Handling, Warehousing, and Inland Transportation”(Marine Department of Hong Kong, 2006). Port’s

poor performance depends on dwell time, delay surcharge and ignorance by bigger ships due to insufficient

infrastructure.

Container terminals are the main drivers in the operational performance of a port by minimizing ships

turnaround time and subsequently maximizing the terminal throughput (Pallis & Syriopoulos, 2007). The yard

operations in container terminals is the busiest of all the activities in the terminal (PwC, 2013). Container yard

operation’s main aim must be to promptly accommodate ships with minimum waiting time in port and with

maximum use of berth facilities (Ruto & Datche, 2015).

Quay length is other effective parameters in efficiency of container which scale on the ability to handle

more containers per one ship within one quay (Yang et al., 2011). Equipment and machines in quay also

contribute to the performance of the ports as they carry out loading and unloading a container from a truck to a

vessel or unloading a container from a vessel to a truck or the vice versa (Babounia et al., 2016).

Alongside quay, Berths too are involved in improving the efficiency of the port terminals. Berth’s

number and length at a container terminal is one of the most important factor the influences the performance and

efficiency of the port (Yang et al., 2011).

Efficient cargo handling operations and adequate infrastructure helps avoid congestion and are the

indicators of efficient Port Infrastructure which ultimately improves trades and container connectivity of the

international trade(Langen et al., 2007). Improved waterfront system, an advancement of infrastructure,

includes; reduced human intervention by automatization, low documentation time, Reduced cargo dwell time,

Reduced port clearance time, Advanced planning technologies, Ease of statistical data calculations, Enhanced

audit trails etc. (Ducruet et al., 2014).

Port dwell time, another influential factor, refers to the time spend by carriers within the port or its

extension (Liu, 2010). Cargo dwell time another important influential factior, is defined as the “time between

vessel arrival and container exit from the port facilities and less the average dwell time more efficient is the

seaport”(Slack & Comtoise, 2015; Pg. 3). Speed of cargo handling is also very critical as faster ship loading and

unloading will raise the ship calls number. Since, the number of terminals is fixed; slow speed will cause the

ship to occupy the berth longer, which will delay the next vessel calling and causes a negative impression(Jafari

et al., 2013).

Financial indicator is the most important indicator of port performance for most of the ports(UNCAD,

2016). Finance indicators include value added, profit, revenues, return on capital employed and others. Value-

added indicators refer to expenses on labour, depreciation and profit, but they are difficult to measure and

compare because of the diversity of the activities involved.





Figure 1: A hypothetical model for Port Efficiency

Port efficiency is bi-directional as the input and output values towards port performance are equally

influential. Input values are applied for high port performance, while the output values show the port efficiency.

Many port efficiency models are present according to the literature review.

Measuring Port Efficiency

Commonly used approaches to measure/analyze the efficiency/productivity include, stochastic 17

frontier analysis (SFA), Data Envelopment analysis (DEA), Vector error model (VEM), Corrected original least

squares (COLS), Original least squares (OLS)(Beasley, 2017).Stochastic Frontier Analysis is a “parametric

and stochastic approach to estimate productive efficiency of ports” which is based on the approach that a

terminal is efficient if it produces a maximum output variables for a given inputs variables(Dong et al., 2014).

Secondly, Corrected Original Least Squares (COLS)is also a method of evaluating ports efficiency using the

parametric approach. It uses regression methods by calculating an average line by enclosing the data and

correction of the line position. The corrected line can then be measured against this frontier (Bates & Bates,

2007).Original Least Squares (OLS) Estimation method basically uses a regression model so that the average

line of the data is adjusted(Chun & Keleş, 2010).

However, in this study DEA program was used to estimate production efficiency(Beasley, 2017). In

this approach, the program maps out a production frontier based on information on inputs and outputs. The

Bi-Directional

Output

Values

Input Values





degree of efficiency is assessed by the distance between the observation and the frontier (Anderson, 2017). The

throughput variables are contrasted to check the output performance of the case ports. DEA programs use input

and output weights by as the basis of the efficiency calculation and such units can be classified into efficient and

inefficient organizations. In inefficient units, the values inform about the amount of input or output variables

required to become efficient.

The input variables used in this are; Infrastructure, Quay, Berths, Financial expenses while, the output

variable were; Tonnage, Net Income, Ship Turnaround time, Waiting time and Vessel Calls. The variables were

taken from secondary sources like annual reports of case ports and literatures by researchers and shipping

companies. Benchmarking of the factors can be done manually by applying graphs and comparing the best

performance of the ports in figures. However, benchmarking of the factors cannot be reliable just on figures

because it does not show the efficiency measurement. Thus, DEA program was used so that efficiency of the

case ports be found and then the factors influencing port efficiency could be benchmarked for the inefficient

ports.

Table 2: Table for the DEA of Input variables

DMU Name Tanger Med Algeciras Bay Rotterdam New York-New Jersey

Berth Length (meters) 632.4 2124 40000 15000

Berth Nos. 4 8 23 3

Quay Length (meters) 1600 600 89000 30000

Quay Nos. 2 2 12 4

Expenses ($) 433433 2909608 1946974 2900652

Channel (meters) 18 15.2 12.2 13.7

Cargo Pier (meters) 18 12.2 9.1 13.7

Anchorage (meters) 18 23.2 13.7 13.7

Oil Terminal (meters) 18 15.2 10 15.2

Table 3: Table for DEA of output variables

DMU Name TangerMed Algeciras Bay Rotterdam New York-New Jersey

Income ($) 527906 941905 733035 737005

Average Turnaround time (Days) 2.5 1 0.5 1.5

Average Waiting time (Hours) 14 9 7 10

Average Vessel calls 51 160 72 62

Exports Tonnage (million TEU) 0.92 12.5 5.4 10.6

Import Tonnage (million TEU) 14 39.5 5.7 26.1

Container handled (million

Tonnes) 3 4.52 12.23 3.7

Liquid Bulk (million Tonnes) 15.5 25.32 209.4 5.05

Dry bulk (million Tonnes) 2.1 1.6 82.6 112.01

General Cargo (million Tonnes) 111 61.16 6.9 38

(Note: DMU stands for decision making units and the Efficient Input/output Targets are the variables(Ray,

2008); The ship turnaround time of the ports were taken from Ducruet et al., 2014)

After the analysis of the performance models of the case ports and subsequent literatures it was found

that the benchmarked values for the port’s operations are; average turnaround time, average waiting time,

tonnage handled and average vessel calls.However, there are other important benchmarking values; dwell time,

crane moves per hour, time spent at anchorage and others could not be evaluated due to lack of data, hence, the

above mentioned factors in table 2 and 3 has been analyzed.

Operational Efficiency of the Case ports

In this section, the performance and the efficiency of the case ports will be interpreted and analyzed.

Each output variable was individually analyzed against similar set of input variables. Firstly Ship turnaround

time was evaluated against similar input variables; quay and berth, infrastructure, similarly the other output

variables were also evaluated accordingly to their input variable.

Contrasting Port Attributes

Expenses and Income ratio of Port Finances

From Figure 2 it can be interpreted that Tanger Med Port is the only one that has surpassed the

expense-income ratio (El Imrani & Babounia, 2016). This might implicate that the Tanger Med port has most

efficiently used their resources. While others had low income, might be because they did not focus on financial

performance.





Figure 2: Contrasting the Expenses and Income ratio of Port Finances

Average Turn-around time of Ships

From the Figure 3 it can be seen that the Port of Rotterdam (0.5days) had the least average ship

turnaround time (Ducruet et al., 2014). The accepted benchmarked value for the average turnaround time of the

case ports was 1.5 days. This means that the Port of Rotterdam, Port of Algeciras Bay and New York-New

Jersey port efficiently maintains its terminal operations.

Figure 3: Benchmarking the Average Turnaround time of Ships

Average ship waiting time

From the Figure 4 it can be interpreted that the Ports of Rotterdam (7hrs), Algeciras Bay (9hrs) and

NY-NJ (10hrs) has low ship waiting time. The accepted benchmark for the average ship waiting time from the

time of call to exit port was 9,25 hours. Tanger Med has high waiting time For two reasons, firstly might be

because they have low efficient infrastructure (Ducruet et al., 2014),or, to take advantage of the infrastructure,

one of the points that is gaining in quality in Morocco, it is necessary that the reforms of “soft” components

accompany (Babounia A., El Imrani O., 2017).





Figure 4: Benchmarking the Average Ship waiting time

Average vessel call

The Figure 5 represents the average vessel calls by shippers and cargo carriers to the sample ports. The

accepted benchmark value for the case port vessel calls was 86. Thus from the graph it can be seen that the Port

of Algeciras Bay had the most number of vessel calls (160) and also shows validity to the results from the fig.3

and fig.2 that low turnaround time and low waiting time influences the number of vessel calls and carrier

preferences (Ducruet et al., 2014; Slack & Comtoise, 2015).

Figure 5: Benchmarking the Average Vessel calls

Tonnage handled

The figure 6 shows the representation of the tonnage handled in various sectors of the terminal

operations. Tanger Med port handled the most general cargo (111 m tonnes), while the dry bulk was most

handled by NY-NJ Port (112.01 m tonnes) and Liquid Bulk was most handled by Port of Rotterdam (209.4 m

tonnes). However, this tonnage cannot be suggested for the most efficiently handled ports, but mainly because

of the hinterland connectivity and market demand. However, the export-import tonnage can be benchmarked for

the most efficient operational port, because lesser the time for ship turnaround more the tonnage for import and

export loading and unloading. Port of Algeciras Bay had handled the most amounts of imports (39.5 m TEU)

and exports (12.5 m TEU). Thus, it again shows higher the efficiency of port and ship operations, larger is the

handling of export-import tonnage (Jafari et al., 2013).





Figure 6: Graphical representation of tonnage handled

Port Efficiency Analysis

Data Envelopment Analysis (DEA) which shows either constant returns to scale (CRS) or variable

returns to scale (VRS) on the basis of the equality of the input-output bundle and identical radial measures of

technical efficiency (Proudlove, 2017). Using this tool the variables were put in MS Excel and DOS command

DEA software and the results were run (Coelli, 2008). Below is the results table from the DEA tool of efficiency

analysis of the benchmark values of the case ports. This tool was used specifically to benchmark the port’s

efficiency by input oriented measures for every input and output variable (Mohammadi et al., 2016). Radial

models can be employed based on input or output discretionary or controllable input-oriented or output-oriented.

This is an input oriented data envelopment analysis. Radial efficiency means “that a proportional input

reduction or a proportional output augmentation is the main concern in assessing the efficiency of the

DMUs”(Fukuyama, 2014; Pg. 1968).

Input oriented efficiency value against Average Ship Turnaround time

This is the CRS DEA result for input oriented efficiency benchmarking. Here the input variables were

Berth Length, No. of Berths, Quay Length and No. of Quays and the output variable was Ship’s turnaround

time. From the Table 4 it can be seen that all the variables are technically efficient for Tanger Med Port and

Algeciras Bay Port, as the efficiency score is equal to one. While for Rotterdam Port (0.035) and New York-

New Jersey Port (0.800) there was inefficiency of the input-output ratio. This means that for Port of Rotterdam

have to do additional improvement by decreasing inputs of 96.5% to become efficient (Port of Rotterdam,

2016b). Again for NY-NJ Port they have to increase output and decrease input by 20% to become efficient

(JOC, 2017).

Table 4: Table for Input oriented efficiency value against Average Ship Turnaround time

Input Oriented CRS DEA

Case Ports or DMU Input Oriented Efficiency Value

Tanger Med 1.000

Algeciras Bay 1.000

Rotterdam 0.035

New York-New Jersey (NYNJ) 0.800

Mean 0.709

Input Slacks against Average Ship turnaround time

Table 5 shows the results of the Input Slacks. In Data Envelopment Analysis, slacks show a unique

value that can be explained on the basis of the excess input and minimal output even after proportional slack

change(Coelli, 2008). However, the role of slacks is in the context of radial measures of efficiency. Tanger Med

0.9212.5

5.410.614

39.5

5.7

26.1

3 4.5212.23

3.715.5

25.32

209.4

5.052.1 1.6

82.6

112.01111

61.16

6.9

38

0

50

100

150

200

250

Tanger Med Algeciras Bay Rotterdam New York-New Jersey

Exports Tonnage (m TEU)

Import Tonnage (m TEU)

Container handled (m TEU)

Liquid Bulk (m tonnes)

Dry bulk (m tonnes)

General Cargo (m tonnes)





and Algeciras port had no slacks as they have an input efficiency value of 1. While Rotterdam and NYNJ

showed slack values for the input variables berth length and quay length. The amount of the excess input that

exists even after the proportional change for Rotterdam Port were found to be 1264.824 (berth length) and

23040.000 (Quay length), while for NYNJ Port 11620.560 (berth length) and 6453.913 (Quay length). Thus, the

Rotterdam port can either lower their input or they can increase their output by 1264.824 (berth length) and

23040.000 (Quay length), while the same can be performed by New York-New Jersey Port by 11620.560 (berth

length) and 6453.913 (Quay length). Slack values lead to another assessment result that is the reference set.

Reference set or also known as peers are the set of efficient units from which an inefficient unit’s inefficiency

can be determined (Anderson, 2017). The peers or reference unit for Rotterdam and NYNJ Ports is Tanger Med

Port. This also means that the low efficiency ports, Rotterdam and NYNJ do not have the best practice and their

benchmark is Tanger Med port which has the best efficiency in case of Berth and Quay operations.

Table 5: Table for Input Slacks against Average Ship turnaround time

Summary of Input Slacks

Case Ports or DMU Berth Length No. of Berths Quay Length No. of Quays

Tanger Med 0.000 0.000 0.000 0.000

Algeciras Bay 0.000 0.000 0.000 0.000

Rotterdam 1264.824 0.000 2775.652 0.017

New York-New Jersey

(NYNJ)

11620.560 0.000 23040.000 2.000

Mean 3221.346 0.000 6453.913 0.504

Input oriented efficiency against Average Ship waiting time and Input Slacks

In this table 6 the input variables were again berth length, no. of berths, quay length and no. of quays,

while the output variable was ship’s waiting time. From the efficiency test it was seen that Tanger med and

Algeciras ports had high efficiency with value of 1. However, Rotterdam (0.087) and NYNJ (0.952) had low

efficiency of port performance. Thus, Rotterdam and NYNJ has to either decrease the input or increase the

output by 91.3% and 5% respectively to become efficient (JOC, 2017; Port of Rotterdam, 2016b). Again, by the

evaluation of input slacks, Rotterdam port can either lower their input or they can increase their output by

3162.061 (berth length) and 6939.130 (Quay length), while the same can be performed by New York-New

Jersey Port by 13834.000 (berth length) and 8591.925 (Quay length) and 2.381 (No. of Quays). The peers or

reference unit for Rotterdam and NYNJ Ports is Tanger Med Port.

Table 6: Table for Input oriented efficiency against Average Ship waiting time and Input Slacks

DMU Input Oriented

Efficiency Value


Berth

Length

No. of Berths Quay Length No. of

Quays

Tanger Med 1.000 0.000 0.000 0.000 0.000

Algeciras Bay 1.000 0.000 0.000 0.000 0.000

Rotterdam 0.087 3162.061 0.000 6939.130 0.000

New York-New

Jersey (NYNJ)

0.952 13834.000 0.000 8591.925 2.381

Mean 0.760 4249.015 0.000 8591.925 0.606

Input oriented efficiency against Average Vessel Calls and Input Slacks

The table 7 shows the efficiency for the input variables berth length, no. of berths, quay length and no.

of quays against output variable Vessel calls. Evaluation for the efficiency of the ports showed that Tanger med,

Algeciras and New York-New Jersey ports had high efficiency with value of 1. The output variable was the

average of calls from every permanent and temporary cargo carriers. Rotterdam port although had more than a

hundred carriers with port of calls but the average vessel calls on monthly basis was low, thus showing an

inefficiency of 0.155. However, the inefficiency can be tackled by increasing their output by 85% (Port of

Rotterdam, 2016b). By the evaluation of input slacks, Rotterdam port can either lower their input or they can

increase their output by 1021.230 (berth length) and 4659.104 (Quay length). In this case the reference unit for

the Rotterdam port is Algeciras port and NYNJ port.

Table 7: Table for Input oriented efficiency against Average Vessel Calls and Input Slacks

DMU Input Oriented

Efficiency Value


Berth Length No. of

Berths

Quay

Length

No. of

Quays

Tanger Med 1.000 0.000 0.000 0.000 0.000

Algeciras Bay 1.000 0.000 0.000 0.000 0.000

Rotterdam 0.155 1021.230 0.000 4659.104 0.000

New York-New

Jersey (NYNJ)

1.000 0.000 0.000 0.000 0.000





Mean 0.789 255.307 0.000 1164.776 0.000

Input oriented efficiency against Import and Export Tonnage and Input slacks

Table 8 shows the results of efficiency test with input variables berth length and quay length, no. of

berths and no. of quays where the Import and Export Tonne age were the output variables. The berth and quay

operations of a port influence the Import and export tonnage. Again, Rotterdam port was found to be the only

inefficient port with 0.114 inefficiency value. However, the Rotterdam port can either decrease the input or

increase the output by 89% to become efficient (Port of Rotterdam, 2016b). Evaluation of slacks showed that,

Rotterdam port had an excess input even after the proportional change of 750.978 and 3426.146 for berth and

quay length respectively. Thus, Rotterdam does not have the best practice and its benchmark ports are Algeciras

and NYNJ port which has the best efficiency in case of Berth and Quay operations against export-import

tonnage.

Table 8: Table for input efficiency against Import and Export Tonnage and Input slacks

DMU Input Oriented

Efficiency Value


Berth

Length

No. of

Berths

Quay

Length

No. of

Quays

Tanger Med 1.000 0.000 0.000 0.000 0.000

Algeciras Bay 1.000 0.000 0.000 0.000 0.000

Rotterdam 0.114 750.978 0.000 3426.146 0.000

New York-New

Jersey (NYNJ)

1.000 0.000 0.000 0.000 0.000

Mean 0.779 187.744 0.000 856.537 0.000

Input oriented efficiency against Container tonnage and input slacks

The input variables were the input variables berth length and quay length, no. of berths and no. of

quays whereas the Container handled are the output variables.. Since the input variables contribute to part of

port infrastructure and port infrastructure influences container handling. Only Tanger med had inefficiency of

0.105, but can improve it efficiency by increasing their output or lowering input by 89.5% (Imrani & Babounia,

2016). Its reference point or peer value is that of Port of Rotterdam and can benchmark its values for high

efficiency. However, according to slacks value, Tanger med can decrease input or increase output by 0.157

(berth nos.) and 0.481 (quay length) to become efficient.

Table 9: Table for efficiency against Container tonnage and input slacks

DMU Input Oriented

Efficiency Value


Berth

Length

No. of Berths Quay

Length

No. of Quays

Tanger Med 0.105 0.000 0.157 0.481 0.000

Algeciras Bay 1.000 0.000 0.000 0.000 0.000

Rotterdam 1.000 0.000 0.000 0.000 0.000

New York-New

Jersey (NYNJ)

1.000 0.000 0.000 0.000 0.000

Mean 0.776 0.000 0.039 0.120 0.000


In this DEA evaluation the input variables were Channel depth, Cargo pier length, Anchorage depth

and Oil terminal length against the output variable of Container handled. Since the input variables contribute to

port infrastructure and port structure influences container throughput. From the table it can be seen that only

Rotterdam port shows efficiency in case of container throughput, while Tanger med (0.187), Algeciras (0.297)

and NYNJ (0.303) shows inefficiency. Thus, Tanger med, Algeciras and NYNJ can become efficient by either

decreasing input or increasing output by 82%, 71% and 70% respectively (El Imrani & Babounia, 2016; JOC,

2017; Parola et al., 2016). Moreover, the slacks value shows that the ports Tanger med could either lower their

input or they can increase their output by 0.368 (berth length), 1.128 (berth nos.) and 0.908 (quay nos.).

Algeciras Bay can also improve their slacks by 0.256 (berth nos.), 1.819 (quay length) and 0.813 (quay nos.)

and for New York-New Jersey port 0.454 (berth length), 1.392 (berth nos.) and 1.573 (quay nos.). The ports

have their peer as Port of Rotterdam and benchmark it for the best practices towards port performance.





Table 10: Table for efficiency test against container tonnage and input slacks

DMU Input Oriented

Efficiency Value


Channel Depth Cargo Pier Anchorage Oil

Terminal

Tanger Med 0.187 0.368 1.128 0.000 0.908

Algeciras Bay 0.297 0.000 0.256 1.819 0.813

Rotterdam 1.000 0.000 0.000 0.000 0.000

New York-New

Jersey (NYNJ)

0.303 0.454 1.392 0.000 1.573

Mean 0.446 0.205 0.694 0.455 0.823


In this analysis the input variables were again Channel depth, Cargo pier length, Anchorage depth and

Oil terminal length against the output variable of dry, oil and general cargo bulks handled by the port. From the

table it can be seen that only Algeciras port has 0.866 of inefficiency and can become efficient by increasing

output by 14% (Parola et al., 2016). Its reference port is Port of Rotterdam and can bench mark its values for

best practices. Moreover, slacks value shows that Algeciras Bay could increase their output by 2.347 (channel),

9.151 (anchorage) and 2.524 (oil terminal) to become efficient.

Table 11: Table for input efficiency against cargo bulks (General, dry and oil) and Input slacks

DMU Input Oriented

Efficiency Value


Channel

depth

Cargo Pier Anchorage Oil

Terminal

Tanger Med 1.000 0.000 0.000 0.000 0.000

Algeciras Bay 0.866 2.347 0.000 9.151 2.524

Rotterdam 1.000 0.000 0.000 0.000 0.000

New York-New

Jersey (NYNJ)

1.000 0.000 0.000 0.000 0.000

Mean 0.966 0.587 0.000 2.288 0.631

Input oriented efficiency against Export-import tonnage and input slacks

In this analysis the input variable were Channel depth, Cargo pier length, Anchorage depth and Oil

terminal length against the output variable of Export and Import tonnage. Mooring is very important in ports and

is done on the berths of the ports. This means that port infrastructure again influences the Export and Import

tonnage.Tanger med and Rotterdam ports showed low efficiency of 0.419 and 0.683 respectively, but can

increase their efficiency by increasing output by 59% and 32% respectively (El Imrani & Babounia, 2016; Port

of Rotterdam, 2016b). However, slacks value shows that Tanger med can decrease their excess input or increase

their minimal output by 0.627 (channel) and 0.862 (cargo pier), while Rotterdam by 0.670 (channel) and 0.564

(cargo pier) to increase efficiency. Both the ports can follow the peer or benchmark values of Algeciras port.

Table 12: Table for efficiency against Export-import tonnage and input slacks

DMU Input Oriented

Efficiency Value


Channel

depth

Cargo

Pier

Anchorage Oil

Terminal

Tanger Med 0.419 0.627 0.862 0.000 0.000

Algeciras Bay 1.000 0.000 0.000 0.000 0.000

Rotterdam 0.683 0.670 0.564 0.000 0.000

New York-New

Jersey (NYNJ)

1.000 0.000 0.000 0.000 0.000

Mean 0.775 0.574 0.357 0.000 0.000

Input oriented efficiency against vessel calls and input slacks

Here the output variable was Average vessel calls and contrasted against Port infrastructure because

many time a vessel may or may not make port of call due to inefficient port infrastructure. Thus, on evaluation it

was seen that Tanger med (0.411), Rotterdam (0.762) and NYNJ (0.656) were inefficient. However, the ports

can either decrease the input or increase the output by 59%, 24% and 35% respectively for Tanger med,

Rotterdam and NYNJ (Imrani & Babounia, 2016; JOC, 2017; Port of Rotterdam, 2016b). However, slacks

values show that Tanger med can decrease their excess input or increase their minimal output by 2.550

(Channel), 3.506 (Cargo Pier) and 2.550 (Oil Terminal); for Rotterdam, 2.457 (Channel), 1.445 (Cargo Pier) and

0.780 (Oil Terminal) and for New York-New Jersey, 3.100 (Channel), 4.262 (Cargo Pier) and 4.084 (Oil

Terminal). The peer or benchmark value for best practice is Algeciras Bay Port.





Table 13: Table for efficiency against vessel calls and input slacks

DMU Input Oriented

Efficiency Value


Channel

depth

Cargo Pier Anchorage Oil

Terminal

Tanger Med 0.411 2.550 3.506 0.000 2.550

Algeciras Bay 1.000 0.000 0.000 0.000 0.000

Rotterdam 0.762 2.457 1.445 0.000 0.780

New York-New

Jersey (NYNJ)

0.656 3.100 4.262 0.000 4.084

Mean 0.707 2.027 2.303 0.000 1.854

Benchmarking Ports

Based on the data envelopment analysis (DEA) for benchmarking and efficiency of the case ports a

rank has been suggested on the basis of the most input oriented efficient Ports.

Table 14: Ranks on the basis of the No. of Input Oriented Efficiency counts

Rank Port No. of Input Oriented Efficiency counts

1 Port of Algeciras Bay, Spain 7

2 Port of Tanger Med, Morocco

Port of New York-New Jersey, USA

5

3 Port of Rotterdam 3

Further, the benchmarked values for each of the ports individually are presented in the table below.

Table 15: Table for characterizing the benchmarked variables for very port and their reference efficient port

Ports Benchmarked Variables Reference Port

Input Output

Port of Tanger

Med

Quay and Berth Infrastructure Container Tonnage Rotterdam

Overall Port Infrastructure

(Channel depth, Cargo pier length,

Anchorage depth and Oil terminal

length)

Container Tonnage Rotterdam

Export-Import Tonnage Algeciras Bay

Vessel calls Algeciras Bay

Port of

Algeciras Bay




length)

Bulk Tonnage Rotterdam


Port of

Rotterdam

Quay and Berth Infrastructure Ship Turnaround time Tanger med

Ship waiting time Tanger med

Vessel calls Algeciras and NYNJ

Import and Export

Tonnage

Algeciras and NYNJ




length)

Export-Import Tonnage Algeciras Bay


Port of New

York-New

Jersey

Quay and Berth Infrastructure Ship Turnaround time Tanger med

Ship waiting time Tanger med




length)



From the above table 15, it can be interpreted that for Tanger med port, their inefficiency of port lies in

infrastructure against Container tonnage, Vessel calls and Import-Export tonnage, but to improve efficiency can

follow the benchmark variables of Port Rotterdam and Algeciras Bay respectively. Similarly, for New York-

New Jersey port, their inefficiency of port lies in infrastructure against Ship turnaround time, Ship waiting time,

vessel calls and container tonnage, but can improve by benchmarking the input variables of Tanger med,

Algeciras and Rotterdam respectively.

III. DISCUSSIONS AND CONCLUSIONS Port efficiency is the measure to check how efficient is the port’s operation and a ratio between output

and input variables of the ports performance. From the analysis of efficiency analysis and benchmarking

variables, it can be said that even if a port may invest in excess to increase the performance and competitiveness

with other ports of the country or the world they may not equally be efficiently working. The case can be





compared with Port of Rotterdam that, even if from the current data collected showed manually that Port of

Rotterdam had the best performance compared to the other three ports but is not equally efficient as the lowest

performing port, Port of Tanger med. This shows relevance with the reports from Port of Rotterdam where they

mention a low port throughput. The ports of NYNJ and Algeciras Bay too showed that even with low input for

high performance, the ports were working efficiently. From the analysis it can be said that the most important

factors that influence the port efficiency are port infrastructure and terminal operations. Other than these factors,

dwell time, crane moves per hour and time spent at anchorage, too contribute to port efficiency. Thus, the

benchmarked variables found are; Quay and Berth Infrastructure, Port Operation and Port infrastructure. The

benchmarking values of the Rotterdam port is container and bulk tonnage which also means that it can follow or

refer Algeciras and Tanger med port to achieve achiement. Algeciras port performance model benchmarked for

Vessel call and Export-Import Tonnage and can be followed by Rotterdam, Tanger med and New York-New

Jersey port. Tanger med port performance model benchmarked for ship turnaround time and waiting time can be

followed by Rotterdam and New York-New Jersey port.

It is recommended that the case ports with inefficient values for respective input and output variables

can follow the reference benchmarked performance models of the case ports. Tanger med port can follow the

performance model of Rotterdam and Algeciras port for Container, Bulk tonnage, Vessel call and Export-Import

Tonnage. Likewise the other ports should follow the benchmarked performance models for efficiency elevation.

Limitations of the study and directions for future research

The present study focused on only comparative analysis as the benchmark research tool to analyze the

ports performance.

This research can be extended to a number of directions that are of substantial importance and provide

important value addition to the current stock of knowledge.

The present study focused on only secondary sources to conduct comparative analysis. An actual focus

group response can be analyzed by developing a qualitative questionnaire and conducting interviews of the

various department managers of the ports where questions asked to them will try to find solutions for

improvement in the ports performance.

We mentioned that this article was funded by the authors.

IV. REFERENCES[1] Anderson, T. (2017). A Data Envelopment Analysis. Retrieved from

http://www.emp.pdx.edu/dea/homedea.html#Returns_to_Scale

[2] Babounia, A.,El Imrani, O. , & Azougagh, K. (2016). A study on improving of port performance to minimize logistics costs of the grain transiting by the tangier med port ISSN : 2278-6236. International Journal of Advanced Research in Management and

Social Sciences, 5(10), 17–32.

[3] Babounia, A.,El Imrani, (2017). Le rôle de l’indice de performance logistique dans le commerce international au Maroc: étude comparative, PROLOG 2017, La Rochelle, France.

[4] Bates, D., & Bates, D. (2007). Comparing Least Squares Calculations. Most, 1–6.

[5] Beasley, J. E. (2017). Data envelopment analysis. Retrieved from http://people.brunel.ac.uk/~mastjjb/jeb/or/dea.html [6] Bozuwa, J., Jorna, R., Recagno, V., & Zografos, K. (2012). BE LOGIC: Benchmarking Logistics Chains. Procedia - Social and

Behavioral Sciences, 48, 2422–2432. https://doi.org/10.1016/j.sbspro.2012.06.1213

[7] Chin-Shan, Lu. Kuo-Chung, S. & Chi-Chang, L. (2016). Examining sustainability performance at ports: port managers’ perspectives on developing sustainable supply chains. Maritime Policy & Management, 43(08), 909-927.

[8] Chun, H., & Keleş, S. (2010). Sparse partial least squares regression for simultaneous dimension reduction and variable

selection. Journal of the Royal Statistical Society. Series B: Statistical Methodology, 72(1), 3–25. https://doi.org/10.1111/j.1467-

9868.2009.00723.x

[9] Coelli, T. J. (2008). A Guide to DEAP Version 2.1: A Data Envelopment Analysis (Computer) Program. CEPA Working Papers,

1–50. Retrieved from https://absalon.itslearning.com/data/ku/103018/publications/coelli96.pdf [10] Dong, Y., Hamilton, R., & Tippett, M. (2014). Cost efficiency of the Chinese banking sector: A comparison of stochastic frontier

analysis and data envelopment analysis. Economic Modelling, 36, 298–308. https://doi.org/10.1016/j.econmod.2013.09.042

[11] Ducruet, C., Itoh, H., & Merk, O. (2014). Time Efficiency at World Container Ports. International Transport Forum Discussion Papers, (2014–8). Retrieved from

www.internationaltransportforum.org/jtrc/DiscussionPapers/jtrcpapers.html%5Cnwww.internationaltransportforum.org

[12] El Imrani, O., & Babounia, A. (2016). Tangier Med Port : What role for the Moroccan Economy and the International Trade ? International Journal of Research in Management, Economics and Commerce, (7), 73–81.

[13] Fukuyama, H. (2014). Radial Efficiency Measures in Data Envelopment Analysis. Encyclopedia of Business Analytics and

Optimization, 1967–1976. https://doi.org/10.4018/978-1-4666-5202-6.ch177 [14] Goodwin, E. (2015). Annual Report 2015. Eurokai, 147. https://doi.org/10.1108/eb055690

[15] Jafari, H., Saeidi, N., Noshadi, E., & Hallafi, H. R. (2013). An Empirical Study of Factors Affecting Reduction of Performance in Container Handling Operation. Journal of Asian Business Strategy, 3(12), 330–339.

[16] Jasmine Siu, L L. , Theo, N. ( 2014). The Greening of Ports: A Comparison of Port Management Tools Used by Leading Ports in

Asia and Europe. Transport Reviews, 32 (2), 169-189. [17] JOC. (2017). Port Authority Of New York & New Jersey. Retrieved from

http://www.joc.com/port?news/us?ports/port?authority?new?york?new?jersey

[18] Langen, P. de, Michiel Nijdam, & Horst, M. van der. (2007). New indicators to measure port performance. Journal of Maritime Research, IV(1), 23–36. Retrieved from

http://www.researchgate.net/publication/28199982_New_indicators_to_measure_port_performance

[19] Liu, Q. (2010). Efficiency Analysis of Container Ports and Terminals, 206. Retrieved from http://eprints.ucl.ac.uk/19215/





[20] Marine Department of Hong Kong. (2006). Port Bechmaking for Assesing Hong Kong’s Maritime Service and Associated Cost

With Other Major International Ports. Marine Department: Planning, Development and Port Security Branch, (December), 1–51.

[21] Michele, A. (2015). Corporate responsibility and value creation in the port sector. International Journal of Logistics Research

and Applications, 18(3), 291-311. [22] Mohammadi, S., Mirdehghan, S. M., & Jahanshahloo, G. (2016). Finding the Most Preferred Decision-Making Unit in Data

Envelopment Analysis, 2016.

[23] Mpogolo, I. A. (2013). Efficiency analysis on operational performance of container terminals for seaports: a case of dar es salaam port. Open University of Tanzania.

[24] Nyema, S. M. (2014). Factors Influencing Container Terminals Efficiency: a Case Study of Mombasa Entry Port. European

Journal of Logistics Purchasing and Supply Chain Management, 2(3), 39–78. [25] Pallis, A. A., & Syriopoulos, T. (2007). Port governance models: Financial evaluation of Greek port restructuring. Transport

Policy, 14(3), 232–246. https://doi.org/10.1016/j.tranpol.2007.03.002

[26] Parola, F., Risitano, M., Ferretti, M., & Panetti, E. (2016). The drivers of port competitiveness: a critical review. Transport Reviews, 1–23. https://doi.org/10.1080/01441647.2016.1231232

[27] PONYNJ. (2014). A Collaborative Effort for a Collective Change. The Port of New York and New Jersey, (June).

[28] PONYNJ. (2016). Financial Statements & Appended Notes. The Port of New York and New Jersey.

[29] Port of Algeciras Bay. (2015). Annual Report. Algeciras Port Authority, 25. https://doi.org/1.1

[30] Port of Algeciras Bay. (2016). Port of the Month : Port of Algeciras. Port of Algeciras Port Authority, 1–8.

[31] Port of Rotterdam. (2016). BUILDING A SUSTAINABLE FUTURE. Port of Rotterdam Authority, 1–16. [32] Port of Rotterdam. (2016). Port of Rotterdam throughput decreased by 3 . 0 %. Retrieved from

https://www.portofrotterdam.com/en/news?and?press?releases/port?of?rotterdam?throughput?decreased?by?30

[33] Port of Rotterdam. (2016). Port of the Month: Port of Rotterdam (The Netherlands). Port of Rotterdam Authority, 1–13. Retrieved from https://www.portofrotterdam.com/en

[34] Proudlove, N. (2017). Using Excel for Basic Data Envelopment Analysis. Retrieved from

https://www.researchgate.net/publication/255708004%0AUsing [35] PwC. (2013). Measures to enhance the efficiency and quality of port services in the EU. European Commission Directorate-

General for Mobility and Transport Unit B3 Ports & Inland Navigation, 1–186. [36] Ray, S. C. (2008). Comparing Input- and Output-Oriented Measures of Technical Efficiency to Determine Local Returns to Scale

in DEA Models. Economics Working Papers, (September), Paper 200837. Retrieved from

http://ideas.repec.org/p/uct/uconnp/2008-37.html [37] Ruto, W. K., & Datche, E. (2015). Logistical factors influencing port performance a case of kenya ports authority ( KPA ).

IJCRR, 7(12), 52–59.

[38] Slack, B., & Comtoise, C. (2015). Ships Time in Port, an international comparison , 1–23. [39] Tanger Med Port Authority. (2014). Port of Tanger Med. Tanger Med Port Authority.

[40] Tanger Med Port Authority. (2016a). ACTIVITE PORTUAIRE GLOBALE 2016. Tanger Med Port Authority.

[41] Tanger Med Port Authority. (2016b). APM Terminals invests 8.5 bMAD in a new terminal in TangerMed II. Tanger Med Group News APM. Retrieved from www.tmsa.ma

[42] UNCAD. (2016). Port Management: Linking Performance Indicators to Strategic Objectives. United Nations Conference On

Trade And Development, 15(9), 46. https://doi.org/10.1057/9781137475770 [43] Vitsounis, T. (2012). Port Performance Measurement in Practice. Port Economics. Retrieved from

http://www.scribd.com/doc/95506907/Port-Performance-Measurement-in-Practice-Vitsounis-Belgrade-1#page=1

[44] Yang, H., Lin, K., Kennedy, O. R., & Ruth, B. (2011). Sea-Port Operational Efficiency: An Evaluation of Five Asian Ports Using Stochastic Frontier Production Function Model. Journal of Service Science and Management, 4(3), 391–399.

https://doi.org/10.4236/jssm.2011.43045

benchmark and competitive analysis of port performances...

Documents